shell脚本，用于查找字符串的第一个和最后一个occurance

Question

我准备了一个shell脚本，用于在一个50节点的hadoop集群中执行以下操作：

• 列出每台服务器中与我的应用程序相关的所有日志文件
• 打印上次修改的时间戳、主机名、文件名
• 根据修改后的时间戳对50个节点中的日志文件进行排序

当前输出格式为：

2016-07-11-01:06 server1 MY_APPLICATION-worker-6701.log.6.gz
2016-07-12-05:23 server1 MY_APPLICATION-worker-6701.log.7.gz
2016-07-13-08:38 server2 MY_APPLICATION-worker-6701.log
2016-07-13-10:38 server3 MY_APPLICATION-worker-6701.log.out
2016-07-13-10:38 server2 MY_APPLICATION-worker-6701.log.err
2016-07-13-10:38 server5 MY_APPLICATION-worker-6701.log
2016-07-15-10:22 server4 MY_APPLICATION-worker-6703.log.out
2016-07-15-10:22 server3 MY_APPLICATION-worker-6703.log.err
2016-07-15-10:22 server2 MY_APPLICATION-worker-6703.log

totallogs=""
for server in $(cat all-hadoop-cluster-servers.txt); do
    logs1="$(ssh user_id@$server 'ls /var/log/hadoop/storm/ -ltr --time-style="+%Y-%m-%d-%H:%M" | grep MY_APPLICATION | awk  -v host=$HOSTNAME "{print $6, host, $7}"' )"
    if [ -z "${logs1}"  ]; then
        continue
    else
        logs1+="n"
        totallogs+=$logs1
    fi  
done
for el in "${totallogs[@]}"
do
    printf "$el"
done | sort

如何在每个日志文件中查找"唯一ID"的第一次出现和"唯一ID"的最后一次出现以及上述输出。

预期输出格式为：

time_stamp主机名文件名第一个唯一ID最后一个唯一ID

2016-07-11-01:06 server1 MY_APPLICATION-worker-6701.log.6.gz    1467005065878   1467105065877
2016-07-12-05:23 server1 MY_APPLICATION-worker-6701.log.7.gz    1467105065878   1467205065860
2016-07-13-08:38 server2 MY_APPLICATION-worker-6701.log         1467205065861   1467305065852
2016-07-13-10:38 server3 MY_APPLICATION-worker-6701.log.out     
2016-07-13-10:38 server2 MY_APPLICATION-worker-6701.log.err     
2016-07-13-10:38 server5 MY_APPLICATION-worker-6701.log         1467305065853   1467405065844
2016-07-15-10:22 server4 MY_APPLICATION-worker-6703.log.out     
2016-07-15-10:22 server3 MY_APPLICATION-worker-6703.log.err     
2016-07-15-10:22 server2 MY_APPLICATION-worker-6703.log         1467405065845   1467505065853

示例日志文件：

DEBUG | 2008-09-06 10:51:44,848 | unique-ID >>>>>> 1467205065861
DEBUG | 2008-09-06 10:51:44,817 | DefaultBeanDefinitionDocumentReader.java | 86 | Loading bean definitions
DEBUG | 2008-09-06 10:51:44,848 | AbstractBeanDefinitionReader.java | 185 | Loaded 5 bean definitions from location pattern [samContext.xml]
INFO | 2008-09-06 10:51:44,848 | XmlBeanDefinitionReader.java | 323 | Loading XML bean definitions from class path resource [tmfContext.xml]
DEBUG | 2008-09-06 10:51:44,848 | DefaultDocumentLoader.java | 72 | Using JAXP provider [com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl]
DEBUG | 2008-09-06 10:51:44,848 | BeansDtdResolver.java | 72 | Found beans DTD [http://www.springframework.org/dtd/spring-beans.dtd] in classpath: spring-beans.dtd
DEBUG | 2008-09-06 10:51:44,848 | unique-ID >>>>>> 1467205065862
DEBUG | 2008-09-06 10:51:44,864 | DefaultBeanDefinitionDocumentReader.java | 86 | Loading bean definitions
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 411 | Finished creating instance of bean 'MS-SQL'
DEBUG | 2008-09-06 10:51:45,458 | DefaultSingletonBeanRegistry.java | 213 | Creating shared instance of singleton bean 'MySQL'
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 383 | Creating instance of bean 'MySQL'
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 459 | Eagerly caching bean 'MySQL' to allow for resolving potential circular references
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 411 | Finished creating instance of bean 'MySQL'
DEBUG | 2008-09-06 10:51:45,458 | DefaultSingletonBeanRegistry.java | 213 | Creating shared instance of singleton bean 'Oracle'
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 383 | Creating instance of bean 'Oracle'
DEBUG | 2008-09-06 10:51:45,458 | AbstractAutowireCapableBeanFactory.java | 459 | Eagerly caching bean 'Oracle' to allow for resolving potential circular references
DEBUG | 2008-09-06 10:51:45,473 | AbstractAutowireCapableBeanFactory.java | 411 | Finished creating instance of bean 'Oracle'
DEBUG | 2008-09-06 10:51:45,473 | DefaultSingletonBeanRegistry.java | 213 | Creating shared instance of singleton bean 'PostgreSQL'
DEBUG | 2008-09-06 10:51:45,473 | AbstractAutowireCapableBeanFactory.java | 383 | Creating instance of bean 'PostgreSQL'
DEBUG | 2008-09-06 10:51:45,473 | AbstractAutowireCapableBeanFactory.java | 459 | Eagerly caching bean 'PostgreSQL' to allow for resolving potential circular references
DEBUG | 2008-09-06 10:51:45,473 | AbstractAutowireCapableBeanFactory.java | 411 | Finished creating instance of bean 'PostgreSQL'
INFO | 2008-09-06 10:51:45,473 | SQLErrorCodesFactory.java | 128 | SQLErrorCodes loaded: [DB2, Derby, H2, HSQL, Informix, MS-SQL, MySQL, Oracle, PostgreSQL, Sybase]
DEBUG | 2008-09-06 10:52:44,817 | DefaultBeanDefinitionDocumentReader.java | 86 | Loading bean definitions
DEBUG | 2008-09-06 10:52:44,848 | unique-ID >>>>>> 1467205065864

Answer 1

grep 'uniqueID' sample_log_file | sed -n '1p;$p'

Answer 2

由于您已经在使用awk，您可以更改awk程序

"{print $6, host, $7}"

至

"{ first=last=""; path="/var/log/hadoop/storm/"$7; while (getline var <path) if (split(var, arr, ">>>>>>") > 1) { if (!first) first=arr[2]; last=arr[2] } print $6, host, $7, "t", first, last }"

以便让它完成工作。